World Congress 2026 Europe • Jul 10, 2026 • Session details

The Limits of LLMs in Real-World Applications

Deivids Vilkinsons , Guillaume Esnou , Laura Moritz , Mariam Hakobyan

Why do most enterprise LLM pilots fail to scale? Discover why poor data and missing guardrails—not the models themselves—are actually derailing your production applications.

Pause
Mute Enter Fullscreen
#1 about 6 min

Shifting from AI hype to enterprise operations

How market limits have shifted organizations toward strict operations and AI accountability.

#2 about 3 min

Where the AI strategy gap appears first

Why real-world edge cases often disrupt workflows built on perfectly planned pilots.

#3 about 2 min

Root causes of underlying AI initiative failures

How messy data and missing operational context cause failures despite powerful artificial intelligence.

#4 about 4 min

Transitioning from demos to real business processes

The challenges of wrapping artificial intelligence applications in critical governance and security.

#5 about 3 min

Determining LLM reliability for production use cases

How acceptable latency and accuracy metrics shift dramatically for safety-critical domains.

#6 about 2 min

Balancing automation with necessary human judgment

Why deterministic processes suit automation while strategic decisions still demand human intervention.

#7 about 1 min

Defining ownership of enterprise AI transformations

Why the stakeholder managing the ultimate business metric should lead the artificial intelligence project.

#8 about 3 min

Measuring practical AI return on investment

Tracking how automated systems shift employee effort from repetitive tasks to high-value outcomes.

#9 about 4 min

Why scaling AI is harder than traditional software

How rapid evolution, soaring compute costs, and isolated agentic context complicate enterprise deployments.

#10 about 3 min

Addressing common enterprise AI misconceptions

Dispelling expectations that artificial intelligence will magically resolve complex integrations or eliminate essential jobs.

#11 about 2 min

Managing compute costs and AI model routing

How organizing model-specific proficiencies and routing logic helps optimize heavy inference costs.

#12 about 3 min

Vibe coding versus established no-code infrastructure

Why businesses benefit more from reliable pre-built system components than generating entire applications through prompts.

#13 about 2 min

Looking ahead at future limitations of AI models

How unformalized tacit knowledge and restrictive physical memory costs will remain persistent scaling challenges.

Matching moments

1:39 min

Assessing the reality of true enterprise AI adoption

Axel Schell Axel Schell +2 · World Congress 2026 Europe

4:32 min

Testing AI limits in enterprise software design

David Tielke David Tielke · World Congress 2026 Europe

6:30 min

Challenges of transitioning to enterprise-wide AI automation

Axel Schell Axel Schell +2 · World Congress 2026 Europe

2:39 min

Identifying the scaling crisis in enterprise AI

Olga Woschitz Olga Woschitz +1 · Europe 2026 Virtual

3:42 min

Why early enterprise AI pilots fail to scale

Dominik Krimpmann Dominik Krimpmann · World Congress 2025

1:50 min

Challenges with language models in enterprise production environments

Ashish Sharma · LIVE